This PDF file contains the front matter associated with SPIE Proceedings Volume 10070 including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

A novel 3D patterned illumination system using an incoherent light source has potential benefits for structured illumination microscopy (SIM) such as lowered intensity requirements and tunable modulating frequency. The illumination system, based on a coherent source, a set of parallel slits and a beam-splitting Fresnel biprism, generates localized interference fringes with a continuously-tunable range of lateral and axial spatial frequency combinations that are not easily accessible using other existing methods of generating structured illumination. Here we present adaptation of this system to a wide-field fluorescence microscope that tests its suitability for SIM imaging. Numerical simulations and experimental data are used to compare theoretical and practical system properties. Results demonstrate that theoretically predicted illumination properties can be used to select system design parameters and accurately produce specific illumination properties at the microscope sample plane. As part of an imaging system, this illumination approach may improve the applicability of super-resolution SIM to a greater variety of samples.

Optical sectioning techniques using two-photon excitation of fluorescent indicators are central to diverse imaging applications. The limitations of the technique are low speed and undesirable specimen agitation. In our design, high-speed axial scanning is carried out by moving a reference objective to axially displace the focal spot without introducing significant spherical aberration and any agitation of the specimen. Further, the system is configured to allow switching between single spot and multiple focal spot remote focusing to allow either high dynamic range or high speed imaging.

New microscopes are needed to help reaching the full potential of 3D organoid culture studies by gathering large quantitative and systematic data over extended periods of time while preserving the integrity of the living sample. In order to reconstruct large volumes while preserving the ability to catch every single cell, we propose new imaging platforms based on lens-free microscopy, a technic which is addressing these needs in the context of 2D cell culture, providing label-free and non-phototoxic acquisition of large datasets. We built lens-free diffractive tomography setups performing multi-angle acquisitions of 3D organoid cultures embedded in Matrigel and developed dedicated 3D holographic reconstruction algorithms based on the Fourier diffraction theorem. Nonetheless, holographic setups do not record the phase of the incident wave front and the biological samples in Petri dish strongly limit the angular coverage. These limitations introduce numerous artefacts in the sample reconstruction. We developed several methods to overcome them, such as multi-wavelength imaging or iterative phase retrieval. The most promising technic currently developed is based on a regularised inverse problem approach directly applied on the 3D volume to reconstruct. 3D reconstructions were performed on several complex samples such as 3D networks or spheroids embedded in capsules with large reconstructed volumes up to ~ 25 mm³ while still being able to identify single cells. To our knowledge, this is the first time that such an inverse problem approach is implemented in the context of lens-free diffractive tomography enabling to reconstruct large fully 3D volumes of unstained biological samples.

In this paper a new, hardware-based solution for extending the depth of field in holographic tomography is presented. The solution is based on a 4f system and an electric, focus-tunable lens, which provides fast, motion-free defocusing of the plane conjugate with the camera, which acquires holograms. The optimum parameters for the required axial scanning are provided for a specific model of a commercially available tunable lens. Then, the quality of the system equipped with the designed module is analyzed and the reconstruction of a standard object (microsphere) scanned by the 4f-based defocusing system is presented. Finally, the result of the increased depth of field in the measurement domain is demonstrated with a reconstruction of a mouse fibroblast cell.

The modular combination of optical microscopes with digital holographic microscopy (DHM) has been proven to be a powerful tool for quantitative live cell imaging. The introduction of condenser and different microscope objectives (MO) simplifies the usage of the technique and makes it easier to measure different kinds of specimens with different magnifications. However, the high flexibility of illumination and imaging also causes variable phase aberrations that need to be eliminated for high resolution quantitative phase imaging. The existent phase aberrations compensation methods either require add additional elements into the reference arm or need specimen free reference areas or separate reference holograms to build up suitable digital phase masks. These inherent requirements make them unpractical for usage with highly variable illumination and imaging systems and prevent on-line monitoring of living cells. In this paper, we present a simple numerical method for phase aberration compensation based on the analysis of holograms in spatial frequency domain with capabilities for on-line quantitative phase imaging. From a single shot off-axis hologram, the whole phase aberration can be eliminated automatically without numerical fitting or pre-knowledge of the setup. The capabilities and robustness for quantitative phase imaging of living cancer cells are demonstrated.

Light-sheet microscopy (LSM) has received great interest for fluorescent imaging applications in biomedicine as it facilitates three-dimensional visualisation of large sample volumes with high spatiotemporal resolution whilst minimising irradiation of, and photo-damage to the specimen. Despite these advantages, LSM can only visualize superficial layers of turbid tissues, such as mammalian neural tissue. Propagation-invariant light modes have played a key role in the development of high-resolution LSM techniques as they overcome the natural divergence of a Gaussian beam, enabling uniform and thin light-sheets over large distances. Most notably, Bessel and Airy beam-based light-sheet imaging modalities have been demonstrated. In the single-photon excitation regime and in lightly scattering specimens, Airy-LSM has given competitive performance with advanced Bessel-LSM techniques. Airy and Bessel beams share the property of self-healing, the ability of the beam to regenerate its transverse beam profile after propagation around an obstacle. Bessel-LSM techniques have been shown to increase the penetration-depth of the illumination into turbid specimens but this effect has been understudied in biologically relevant tissues, particularly for Airy beams. It is expected that Airy-LSM will give a similar enhancement over Gaussian-LSM. In this paper, we report on the comparison of Airy-LSM and Gaussian-LSM imaging modalities within cleared and non-cleared mouse brain tissue. In particular, we examine image quality versus tissue depth by quantitative spatial Fourier analysis of neural structures in virally transduced fluorescent tissue sections, showing a three-fold enhancement at 50 μm depth into non-cleared tissue with Airy-LSM. Complimentary analysis is performed by resolution measurements in bead-injected tissue sections.

Fluorescence microscopy is an important tool in biomedical imaging. An inherent trade-off lies between image quality and photodamage. Recently, we have introduced rescan confocal microscopy (RCM) that improves the lateral resolution of a confocal microscope down to 170 nm. Previously, we have demonstrated that with controlled-light exposure microscopy, spatial control of illumination reduces photodamage without compromising image quality. Here, we show that the combination of these two techniques leads to high resolution imaging with reduced photodamage without compromising image quality. Implementation of spatially-controlled illumination was carried out in RCM using a line scanning-based approach. Illumination is spatially-controlled for every line during imaging with the help of a prediction algorithm that estimates the spatial profile of the fluorescent specimen. The estimation is based on the information available from previously acquired line images. As a proof-of-principle, we show images of N1E-115 neuroblastoma cells, obtained by this new setup with reduced illumination dose, improved resolution and without compromising image quality.

Structured illumination microscopy (SIM) doubles the lateral resolution and produces optically-sectioned images. In SIM, the illumination system is modified in order to illuminate the sample by a structured pattern. Previously, axially-localized high-contrast sinusoidal patterns generated using a slit-prism illumination system based on a Fresnel biprism were investigated. In this contribution, we propose a Wollaston prism to replace the Fresnel biprism and produce the corresponding 3D structured illumination pattern. In this study, both optical elements are illuminated by the light emerging from an axial point source. Our results show that the benefits of using a Wollaston prism instead of a Fresnel biprism are twofold: (1) there is no envelope modulation perturbing the sinusoidal patterns and thereby reducing their visibility, and (2) the region of interference fringes is significantly larger than the one created by the Fresnel biprism.

Single molecule super-resolution microscopy is a powerful tool that enables imaging at sub-diffraction-limit resolution. In this technique, subsets of stochastically photoactivated fluorophores are imaged over a sequence of frames and accurately localized, and the estimated locations are used to construct a high-resolution image of the cellular structures labeled by the fluorophores. Available localization methods typically first determine the regions of the image that contain emitting fluorophores through a process referred to as detection. Then, the locations of the fluorophores are estimated accurately in an estimation step. We propose a novel localization method which combines the detection and estimation steps. The method models the given image as the frequency response of a multi-order system obtained with a balanced state space realization algorithm based on the singular value decomposition of a Hankel matrix, and determines the locations of intensity peaks in the image as the pole locations of the resulting system. The locations of the most significant peaks correspond to the locations of single molecules in the original image. Although the accuracy of the location estimates is reasonably good, we demonstrate that, by using the estimates as the initial conditions for a maximum likelihood estimator, refined estimates can be obtained that have a standard deviation close to the Cramér-Rao lower bound-based limit of accuracy. We validate our method using both simulated and experimental multi-emitter images.

Three-dimensional (3D) single molecule fluorescence microscopy affords the ability to investigate subcellular trafficking at the level of individual molecules. An imaged single molecule trajectory, however, often reveals only limited information about the underlying biological process when insufficient information is available about the organelles and other cellular structures with which the molecule interacts. A new 3D fluorescence microscopy imaging modality is described here that enables the simultaneous imaging of the trajectories of fast-moving molecules and the associated cellular context. The new modality is called remote focusing multifocal plane microscopy (rMUM), as it extends multifocal plane microscopy (MUM) with a remote focusing module. MUM is a modality that uses multiple detectors to image distinct focal planes within the specimen at the same time, and it has been demonstrated to allow the determination of 3D single molecule trajectories with high accuracy. Remote focusing is a method that makes use of two additional objective lenses to enable the acquisition of a z-stack of the specimen without having to move the microscope’s objective lens or sample stage, components which are required by MUM to be fixed in place. rMUM’s remote focusing module thus allows the cellular context to be imaged in the form of z-stacks as the trajectories of molecules or other objects of interest are imaged by MUM. In addition to a description of the modality, a discussion of rMUM data analysis and an example of data acquired using an rMUM setup are provided in this paper.

Cyclic AMP (cAMP) is a ubiquitous second messenger known to differentially regulate many cellular functions. Several lines of evidence suggest that the distribution of cAMP within cells is not uniform. However, to date, no studies have measured the kinetics of 3D cAMP distributions within cells. This is largely due to the low signal-tonoise ratio of FRET-based probes. We previously reported that hyperspectral imaging improves the signal-to-noise ratio of FRET measurements. Here we utilized hyperspectral imaging approaches to measure FRET signals in five dimensions (5D) – three spatial (x, y, z), wavelength (λ), and time (t) – allowing us to visualize cAMP gradients in pulmonary endothelial cells. cAMP levels were measured using a FRET-based sensor (H188) comprised of a cAMP binding domain sandwiched between FRET donor and acceptor - Turquoise and Venus fluorescent proteins. We observed cAMP gradients in response to 0.1 or 1 μM isoproterenol, 0.1 or 1 μM PGE₁, or 50 μM forskolin. Forskolin- and isoproterenol-induced cAMP gradients formed from the apical (high cAMP) to basolateral (low cAMP) face of cells. In contrast, PGE1-induced cAMP gradients originated from both the basolateral and apical faces of cells. Data suggest that 2D (x,y) studies of cAMP compartmentalization may lead to erroneous conclusions about the existence of cAMP gradients, and that 3D (x,y,z) studies are required to assess mechanisms of signaling specificity. Results demonstrate that 5D imaging technologies are powerful tools for measuring biochemical processes in discrete subcellular domains.

Optical imaging modalities are proved to be able to provide images with resolution required to image subcellular particles, however, imaging depth of optical imaging modalities are limited due to strong absorption and scattering. In contrast, ultrasound imaging modalities can provide images deeper in the tissue due to negligible scattering in tissue, but they suffer from poor resolution and contrast. Hybrid imaging modalities such as ultrasound modulated optical tomography (UOT) utilize advantages of both optical and ultrasound imaging modalities. UOT utilizes pressure waves to modulate light with ultrasound frequency that results in a week signal that requires expensive detection equipment. In Contrast, we propose to use acoustic radiation force (ARF) to tag the light that travels through the ultrasound focal spot and generate a stronger signal. Monitoring the changes in the speckle pattern reflects both mechanical and thermal properties of the medium. In this paper we have utilized our model with fixed-particle Monte Carlo to simulate the mean irradiance change (MIC) signal variations due to particle displacement and temperature rise. Results suggest that neglecting the temperature rise for short ultrasound exposure times, the change in the MIC signal reflects the local stiffness of the medium at the ultrasound focal spot and can be utilized to generate the stiffness image of the medium.

Networks of neurons are inherently three-dimensional in nature, whereas conventional imaging methods, such as laser scanning two-photon microscopy, usually provide only fast two-dimensional imaging. Rapid volumetric imaging would however be preferable for imaging neurons. To get a more complete picture of the dynamics of the neuron-to-neuron interactions, we have developed a pseudo-parallelised multi-plane two-photon excitation imaging system through the incorporation of an acousto-optic switching and a remote focusing technique into a resonant scanning microscope. This permits the recording of millisecond scale fluorescence transients of calcium indicators from large populations of neurons upon neural firing events at multiple chosen axial planes in very short time frame. While the remote focusing system offers aberration-free axial scanning over a few hundreds of micrometres of depth, the acousto-optic deflector provides high speed optical switching between different laser beam paths in sub-microsecond timescale which in turn, controls the axial focal plane to be targeted. Here, we report on the development of the high temporal resolution multi-plane targeted microscope and its potential application.

Scanning microscopy methods require movement of the focus in Z coordinates to produce an image of a 3-dimensional volume. In a typical imaging system, the optical setup is kept fixed and either the sample or the objective is translated with a mechanical stage driven by a stepper motor or a piezoelectric element. Mechanical Z scanning is precise, but its slow response and vulnerability to mechanical vibrations and stress make it disadvantageous to image dynamic, time-varying samples such as live cell structures. An alternative method less susceptible to these problems is to change the focal plane using conjugate optics. Deformable mirrors, acousto-optics, and electrically tunable lenses have been experimented with to achieve this goal and have attained very fast and precise Z-scanning without physically moving the sample. Here, we present the use of a liquid lens for fast axial scanning. Liquid lenses have a long functional life, high degree of phase shift, and low sensitivity to mechanical stress. They work on the principle of refraction at a liquid-liquid interface. At the boundary of a polar and an apolar liquid a spherical surface is formed whose curvature can be controlled by adjusting its relative wettability using electro-wetting. We characterize the effects of the lens on attainable Z displacement, beam spectral characteristics, and pulse duration as compared with mechanical scanning.

Angles of polygon scanners have been measured by using rotary encoders, autocollimators or indexing tables. These methods produce precise angle values but require removal of polygon mirror from its motor. For resolving this inconvenience, we introduce a simple angle measurement method by measuring timing jitters of a scanned beam in the time-domain with a high-speed detector and a digitizer while a polygon scanner is rotating at its full speed. Our setup includes a 635 nm wavelength semiconductor laser, a high-speed photodiode, two lenses, and a high-speed digitizer. A polygon scanner with 12 facets were tested with a rotating frequency of near 350 Hz. To detect the signal of the photodiode, we used a high speed digitizer which has a sampling rate of 2Gs/s with 256MB on-board memory. We obtained repeated pulsed sequential photodiode signals for 12 mirror facets of the scanner. Angle variations and their jitters for 12 scanner mirror facets were successfully calculated from measured data. We have repeated same experiments with a photomultiplier tube and compared results with those measured by a photodiode.

A whole rodent brain was imaged using an automated massive histology setup and an Optical Coherence Tomography (OCT) microscope. Thousands of OCT volumetric tiles were acquired, each covering a size of about 2.5x2.5x0.8 mm3 with a sampling resolution of 4.9x4.9x6.5 microns. This paper shows the techniques for reconstruction, attenuation compensation and segmentation of the sliced brains. The tile positions within the mosaic were evaluated using a displacement model of the motorized stage and pairwise coregistration. Volume blending was then performed by solving the 3D Laplace equation, and consecutive slices were assembled using the cross-correlation of their 2D image gradient. This reconstruction algorithm resulted in a 3D map of optical reflectivity for the whole brain at micrometric resolution. OCT tissue slices were then used to estimate the local attenuation coefficient based on a single scattering photon model. The attenuation map obtained exhibits a high contrast for all white matter fibres, regardless of their orientation. The tissue optical attenuation from the intrinsic OCT reflectivity contributes to better white matter tissue segmentation. The combined 3D maps of reflectivity and attenuation is a step toward the study of white matter at a microscopic scale for the whole brain in small animals.

Manipulating the excitation pattern in optical microscopy has led to several super-resolution techniques. Among different patterns, the lateral sinusoidal excitation was used for the first demonstration of structured illumination microscopy (SIM), which provides the fastest SIM acquisition system (based on the number of raw images required) compared to the multi-spot illumination approach. Moreover, 3D patterns that include lateral and axial variations in the illumination have attracted more attention recently as they address resolution enhancement in three dimensions. A threewave (3W) interference technique based on coherent illumination has already been shown to provide super-resolution and optical sectioning in 3D-SIM. In this paper, we investigate a novel tunable technique that creates a 3D pattern from a set of multiple incoherently illuminated parallel slits that act as light sources for a Fresnel biprism. This setup is able to modulate the illumination pattern in the object space both axially and laterally with adjustable modulation frequencies. The 3D forward model for the new system is developed here to consider the effect of the axial modulation due to the 3D patterned illumination. The performance of 3D-SIM based on 3W interference and the tunable system are investigated in simulation and compared based on two different criteria. First, restored images obtained for both 3D-SIM systems using a generalized Wiener filter are compared to determine the effect of the illumination pattern on the reconstruction. Second, the effective frequency response of both systems is studied to determine the axial and lateral resolution enhancement that is obtained in each case.

Structured illumination microscopy (SIM) works well to produce depth information by removing undesired light from out of focus planes within a specimen. However, it generally requires multiple images in order to produce sectioning. We show that, under the right conditions, we can leverage Spatial Frequency Domain Imaging (SFDI) to produce both sectioning and relative depth information about a specimen using only a single image. After processing, we find that the axial resolution is comparable to that of traditional 3 phase SIM. With this technique, we are able to produce real-time videos in-vivo at 21 frames per second. Additionally, no moving parts are needed to produce sectioning at depth.

We propose to couple a single-pixel camera with a photon counting board in order to obtain an inexpensive time-resolved imaging system having a high spatial and temporal resolution. As an alternative to compressive sensing, a wavelet-based adaptive acquisition strategy is employed which allows high compression rates with little degradation of the image quality. The applicability of our approach is demonstrated for fluorescence lifetime imaging. Experimental results obtained by imaging samples embedding several fluorophores are provided. The proposed imaging system with the wavelet-based strategy can be suitable for a microscope in order to perform fluorescence lifetime imaging microscopy measurements.

In this paper, a multi-beam optical coherence tomography (OCT) was used to reconstruct the microvascular image of human skin in vivo with phase resolved Doppler OCT (PRDOCT), phase resolved Doppler variance (PRDV) and speckle variance OCT (svOCT), in which the blood flow image was calculated by averaging the four blood flow images obtained by the four beams. In PRDOCT method, it is difficult to detect the blood flow perpendicular to optical axis of the probe beam for single beam OCT, but the multi-beam scanning method can solve this because the input angles of the four probe beams are slightly different from each other. The proposed method can further improve the signal-to-noise ratio (SNR) of the blood flow signals extracted by the three methods mentioned above.

This paper presents a new method of blood flow imaging in vessels within and underneath the skin. The combination of Laser Speckle Contrast Imaging (LSCI) and Spatial Frequency Domain Imaging (SFDI)/ Structured Illumination Microscopy (SIM) allows one to retrieve a substantial amount of flow information, which is otherwise occluded by the skin. In this article, the experiment consists of illuminating a phantom with coherent light passing through a grid pattern which is modulated at the plane of interest. The speckle pattern is then imaged and processed using a combination of structured illumination sectioning and laser speckle contrast. The whole process is realized using an inexpensive setup consisting of a laser diode (Wavelength = 758 nm), mirrors, lenses, a grid and an 8-bit camera. This advanced, yet easy to implement, technique will allow deep skin imaging to be fast, effective and inexpensive.

We investigated lateral resolution in super-resolution microscopy based on fluorescence depletion, when an annular phase plate is applied to the erase beam. Applying a high NA the objective lens to microscopy, resolution becomes closer to that given by a spiral phase plate.

There is a direct evidence that the radiation doses associated with CT scans are associated with an increase in cancer risk. To reduce the radiation dose and simultaneously maintain the CT reconstruction quality, numerous algorithms have been proposed such as compressive sensing (CS) technique. CS theory asserts that one can recover certain signals and images from far fewer samples or measurements than traditional methods use. In this study, we mainly consider the relationship between the CT reconstruction quality and two undersampled scan types of CS technique, i.e., the sparse-view scan and limited-view scan. The results demonstrate that an appropriate selection of scan type of CS technique can effectively control the radiation dose.

We report a novel microscopy technique, utilizing our previously reported expanded point information content (EPIC) concept [1], to extend the technique into the coherent regime. Preliminary data shows coherent EPIC (CoEPIC) can image reflective samples successfully, and can recover the 3D structure without the need to acquire an image stack at multiple depths. Numerical simulation demonstrates the potential super-resolution capability of CoEPIC.